Since the discovery of penicillin in 1928 the apparent ability of the ever-growing numbers of available antibiotics to treat infections and disease has, until recently, caused a high degree of complacency about the threat of bacterial resistance. This complacency has created a situation where antibiotics are over-prescribed in both hospitals and in the community, and used extensively in animal feeds. The alarming speed with which bacteria have become resistant to microbial agents has meant that there is a very real danger that infections, which were until recently completely controllable, will pose serious threats to human health.
All unicellular bacteria contain a cell wall which is associated with a diverse range of functions, although the major one is that of protecting the cell from lysing under high internal osmotic pressures. The cell wall is composed of peptidoglycan, a rigid mesh of β-1,4-linked carbohydrate polymers covalently cross-linked by peptide chains. The peptidoglycan synthetic pathway is not present in mammalian systems, suggesting that the side-effects associated with such inhibitors could be minimized. Thus the bacterial peptidoglycan biosynthetic pathway presents an opportunity for the development of novel antibacterial agents.
There is a great deal of interest in the substrates of the muramyl pathway and their analogues, and in the synthesis of related compounds that may result in new therapeutics. Tanner and co-workers have recently prepared compounds that inhibit the MurD and MurE enzymes of the muramyl pathway. These non-carbohydrate compounds have the sugar and lactate moieties of a muranic acid-like compound replaced with a five carbon linker unit (Zeng, B., Wong, K. K., Pompliano, D. L., , Reddy, S., and Tanner, M. E., J. Org. Chem. 1998, 63(26), 10081-5; Tanner, M. E., Vaganay, S., van Heijenoort,J., and Blanot, D., J. Org. Chem. 1996, 61(5), 1756-60), and are prepared by standard organic chemistry techniques. They are linear, flexible organic compounds with substituents that resemble those of UDP-MurNAc-pentapeptide (the “Park Nucleotide” (Park, J., J. Biol. Chem. 1952, 194, 877)). One of those compounds in particular was found to be a relatively potent inhibitor of MurE (Zeng, B., Wong, K. K., Pompliano, D. L., Reddy, S., and Tanner, M. E., J. Org. Chem. 1998, 63(26), 10081-5).
In other studies on an analogous phosphinate inhibitor of MurD, it was found that retaining the MurNAc sugar residue, instead of replacing it with a carbon linker unit, increases the potency of the inhibitor by almost two orders of magnitude (Gegnas, L. D., Waddell, S. T., Chabin, R. M., Reddy, S., Wong, K. K., Bioorg. Med. Chem. Lett. 1998, 8, 1643). This suggests that building a library of monosacchatide analogues of the substrates of the muramyl pathway is an attractive proposition for the generation of new therapeutics which target that system.
One approach to the synthesis of such compounds is to make use of biosynthetic techniques, such as that used in preparing labeled versions or analogues of MurNAc from GlcNAc by implementing the MurA and MurB enzymes themselves (Lees, W. J., Benson, T. E., Hogle, J. M., and. Walsh, C. T., Biochemisty 1996, 35(5), 1342-1351).
Chemical methods require protected building blocks, and some well-established chemistry has been implemented, using GlcNAc to yield the benzyl glycoside of N-acetyl-4,6-benzylidenemuramic acid (Jeanloz, R. W., Walker, E., Sinaÿ, P., Carbohydr. Res. 1968, 6, 184). One challenge to the synthesis of such compounds is the alkylation of the C-3 position of the carbohydrate residue. In the natural muramyl system, the MurA and MurB enzymes add what is ultimately a lactate moiety to the C-3 position.
The addition of a lactate moiety at C-3 has been achieved chemically in a process in which the required materials were generated through the intermediate preparation of a nitroalkene sugar (Vega-Perez, et al., Tetrahedron 1999, 55, 9641-9650). An alternative approach is the alkylation of the C-3 hydroxyl with the α-bromide of an appropriately protected propanoic acid to generate the required compound (Iglesias-Guerra, F., Candela, J. I., Bautista, J., Alcudia, F., and Vega-Perez, J. M., Carbobydr. Res. 1999, 316, 71-84).
Having compounds with a lactate moiety, or similar acid, in place at C-3 allowed the addition of amino acids to build the required pentapeptide substituent. This molecule was subsequently converted to the natural substrates for the muramyl enzyme system (Hitchcock, C. N., Eid, J. A., Aikins, M. Z-E., and Blaszczak, L. C., J. Am. Chem. Soc. 1998, 120(8), 1916). In a similar approach the preformed pentapeptide was added as a single unit to yield muramyl products (Ha, S., Chang, E., Lo, M-C., Men, H., Park, P., Ge, M., and Walker, S., J. Am. Chem. Soc. 1999, 121(37), 8415).
Combinatorial chemistry and parallel synthesis have become the methods of choice for the rapid synthesis of a large number of related compounds simultaneously, and this approach has been used to produce libraries of compounds to be screened for biological activity. Sometimes such libraries are focused to test for activity of the compounds so generated towards a particular biological agent or organism, although often large libraries are also prepared in a random fashion. Either way, the intended end result of combinatorial chemistry is the rapid discovery and optimization of leads for the development of new pharmaceuticals.
Despite the obvious advantages of a combinatorial approach to the preparation of compounds for drug discovery, this technique is underexplored in the field of carbohydrate chemistry. This is primarily because of the well-known difficulties associated with the synthesis of carbohydrate compounds. For that reason carbohydrate libraries prepared in the past have tended to be relatively simple. For example, Hindsgaul et al have produced a library of monosaccharide compounds by a combinatorial approach (Ole Hindsgaul, U.S. Pat. No. 5,780,603); however, the variation in the compounds was limited to the glycosidic bond. A glycopeptide library in which mannose residues were decorated with various amino acids has been described, but these were conjugated to the sugar solely through the C-6 position (Tennant-Eyles, R. J., and Fairbanks, A. J., Tetrahedron Asymmetry 1999, 10, 391-401).
Access to greater variation has been attempted by making used of libraries of carbohydrate numetics (Byrgesen, E., Nielsen, J., Willert, M., and Bols, M., Tetrahedron Lett. 1997, 38, 5697-5700 and Lohse, A., Jensen, K. B., and Bols, M., Tetrahedron Lett., 1999, 40, 3033-3036). However, one approach which successfully added greater diversity to monosaccharides was that of Goebel and Ugi (Tetrahedron Lett., 1995, 36(34), 6043-6046) who generated a small library of alkylated glycals by subjecting protected glucals to electrophilic attack and then subsequent reactions. Unfortunately this method is limited by the fact that each starting glucal may give rise to a number of isomeric products.
For these reasons there is particular interest in libraries of aminoglycosides and amino sugars for drug discovery. Some work on such compounds has been published, with Silva and co-workers preparing impressive disaccharide libraries containing glucosamine (Silva, D. J., Wang, H., Allanson, N. M., Jain, R. K., and Sofia, M. J., J. Org. Chem. 1999, 64(16), 5926-5929). However, this library still suffers from the limitation that the variation is limited solely to acylations of the amino group.
More variation, and in fact a three-dimensional diversity, was obtained in the preparation of amino sugars by Sofia and co-workers (Sofia, M. J., Hunter, R., Chan, T. Y., Vaughan, A., Dulina, R., Wang, H., and Gange, D., J. Org. Chem. 1998, 63(9), 2802-2803). This allowed chemical diversity at three combinatorial sites on the sugar residue. Other workers have prepared a library of compounds with four (Wunberg, T., Kallus, C., Opatz, T., Henke, S., Schmidt, W., and Kunz, H., Angew. Chem. Int. Ed. 1998, 37(18), 2503-2505), and five (Kallus, C., Opatz, T., Wunberg, T., Schmidt, W., Henke, S., and Kunz, H., Tetrahedron Lett. 1999, 40, 7783-7786) such sites of functionalization, although these compounds were not amino-sugars.
It will be clearly understood that, although a number of prior art publications are referred to herein, this reference does not constitute an admission that any of these documents forms part of the common general knowledge in the art, in Australia or in any other country.
Hitherto, there have been few attempts to synthesize analogues of the muramyl substrates, particularly those which contain modifications at the anomeric position or at the C-2 nitrogen. The natural substrate and all of the muramyl enzyme intermediates contain exclusively the α-glycosidic diphosphate. Our modeling and design studies with the crystal structure of the MurD enzyme suggest that both the α or β anomeric configuration of many of the compounds proposed in this invention can fit into the active site of this enzyme. We believe that this is the first time that β-glycosides which contain no phosphate groups have been prepared as potential inhibitors of the muramyl enzyme system.
Many of the traditional methods of carbohydrate synthesis have proved to be unsuitable to a combinatorial approach, particularly because modern high-throughput synthetic systems require that procedures to be readily automatable.
The documents cited throughout this application are incorporated into this application by reference.